Hybrid ANN with Invasive Weed Optimization Algorithm, a New Technique for Prediction of Gold and Silver in Zarshuran Gold Deposit, Iran

Because of high cost of drilling and analysis of samples, it needs to predict gold and silvers based on pathfinder such as As, Sb, Cd, Pb and Zn and decrease the cost and time exploration project implementation. In this paper, the model based on a multilayer perceptron artificial neural network (MLP-ANN) optimized by invasive weed optimization algorithm (IWO) to predict of gold and silver in Zarshuran gold deposit, Iran is proposed. The IWO is used to decide the initial weights of the ANN. The ANN-IWO model is trained using an experimental data set to approximate the relation between Sb, Cd, Pb and Zn as inputs and gold and silver as output. Furthermore, the performance of the ANN-IWO model is compared with multiple linear regression (MLR). The results obtained indicate that the ANN-IWO model has strong potential to prediction of gold and silver with high degree of accuracy and robustness.